LDRs as pots?

[Steve Conner]

Oh, I didn't realise they'd been clobbered by the RoHS thing.

Ponder Figures 2 and 6 on the datasheet. To me, these say that they aren't well suited as linear variable resistors at all. For example, looking at Figure 6, you'll see that if you set up a H11F1 as a 10k "resistor", you can't have more than 60mV RMS developed across it before it becomes non-linear.

And from figure 2 you see that the non-linearity is a kind of soft S-curve clipping as it "runs out of current". That's OK when it's used as a series control element, since it just adds compression, but not so pretty when it's used as a shunt to ground...

Also, check out the worst-case dark current spec: 50 microamps at 100 deg.C and 15V!

For these reasons, I've never had much luck using them except as switches.

I'm sure I've written about the H11F1 on this board before, but maybe it wasn't in this thread... You've got me confused now

[TucsonSean]

The nice thing is about them being RoHS compliant is that the price dropped by more than half. They are listing them for $0.43/1000pcs. Much better than the $2+ price they used to go for.

I'll review the datasheet again but I have not given up hope. I have been working on a similar project since 2000. I would like to see a viable alternative to the Vactrol control that Mesa uses.

[Gregg]

Hi,

Some updates.
I transferred the LDR pot from that christmas tree board to a new one.



Instead of LM324 now I'm using LM358. The LDR pot is set up as a regular 1M Gain pot. I tried two configurations: a/ in 2204 type of preamp b/ in a typical Fender style clean channel. I did this by rewiring my DIY JMP-1 preamp as an all tube preamp. because now I have one digital pot free I'm using it to driving the LDR Gain pot. The observations I made:

1/ Concerning distortion

1% Distortion we cannot afford even in a tube amp with a 30K impedance at a 1.4VRMS, not just because thats a lot of distortion for a "clean" sound, but because what we would need for a gain pot would be something like 500K and 2 to 7 VRMS, which would make the distortion so high that only in a "rectifryer" channel would be acceptable.

Well the above maybe right in theory but in practice there's no audible distortion whatsoever /unless you're one of those Hi-Fi guys who by listening a CD can recognize the software used to burn it.../. The clean channel is clean, tight and fat. And no distortion.

2/ Concerning oscillation

The first time there was no audible oscillation in any position /maybe because I was testing it only in a recto style of preamp/. This time however when playing through the clean channel I noticed some. Nothing loud but still annoying and changing tonally when rotating the encoder /dig. pot/. Decided to try some caps across pins 2-1 and 6-7. The first cap I tried - 220pf did the job. Tried several more values without any noticeable effect/improvement. Same result when trying resistors in parallel with the cap.

The next thing to do will be the Tone Stack. I currently have the stock JMP-1 active tone stack but just out of curiosity I want to try the "regular" one . The configuration I intend to use is: CF - flat freq. response fixed TS - buffer /tube or FET/ - regular TS. This is to avoid the high amplitude signals normally passing through the TS.

I hope I have time to record few samples as well.

[Don Moose]

Nifty!

Please keep posting your progress.

[Bruce / Mission Amps]

And by the way... good luck trying to find any H11F1 chips anymore... you might find them in some electronic warehouse but they are obsolete and the word is out ..... they will be in the $2.50 to $3.50 price range each.

[Steve Conner]

Well, like I explained, they were useless for linear gain control in tube amps anyway. Gregg's circuit there uses an op-amp and two LED/LDR combos.

After this thread died, I did some further work on using the H11F1 as a voltage-controlled soft clipper, in an attempt to give tube-like overdrive characteristics to a Class-D power amp. It worked, but no matter what I did, it sounded a bit brittle and trashy. I guess the distortion harmonics from the H11F1 just don't sound that nice. I abandoned that project when the Class-D module blew up on me.

So I guess the H11F1 is really only useful as a switch in low-level stages.

(Gregg, did someone steal your bracket keys?)

[jbefumo]

VTL5Cx Spice model?

I used to use the Clairex CLM-6000 for various projects, and have read that thte VTL5Cx is a more readily available substitute. I was wondering if anyone might point me toward a SPICE model for this component?

Thanks

Joe

[Gregg]

Hi,

As you already know I implemented the LDR pot in a circuit as Gain pot.
Recently I decided to try it as Presence pot. Unfortunately I wasn't able to get the resistance below 2-2.5k although I'm using photocells that on daylight would go under 200 Ohms. Also tried different photocells and VTL5C3/2 , which according to data sheet would go down to 5 Ohms but no matter what I still can't get less than 2-2.5k.

Any ideas what might be the cause for that?

[Don Moose]

The two factors are LED current and time.

Have you tried getting the LED current to, say 95% of the (temperature derated) max?
Then, I think the datasheets specify that the cell doesn't get to Rmin for, like, an hour.

Then there's the temptation to parallel LDRs to get the R lower - just remember that LED's (like other diodes) don't like to be paralleled - they EACH need a series resistor ... though if you have the control voltage headroom, you could put the LEDs in series and the resistors in parallel.

Hope this helps!

[Gregg]

Have you tried getting the LED current to, say 95% of the (temperature derated) max?

Yes, I tried. The same single photocell that would go under 200 Ohms when exposed to daylight or when implemented in a single cell optocoupler wouldn't go down below 2-2.5k when implemented as dual cell optocoupler.
Maybe this has something to do with the cervo/feedback mechanism of the control circuit?

[Koreth]

Wow. What an interesting thread. A lot of cool ideas in this one. Please forgive this act of necromancy, but I figured the question would be better appeneded to this thread than it would be to start a a new one.

It seems that most of the thread is concerned with using LDRs to make voltage dividers to be used as controls, (gain, tonestack, etc) but has anyone thought of using the LDRs to form a simple variable resistor, that one could then use as the plate and cathode resistors of a tube amp stage?

Think about it. You could have all the gain stages in your amp optimized for a beatuiful warm, sparkly clean, then with a MIDI comand to the control circuitry, the load lines and bias points change to what would be ideal for an dirty clean that's dancing on the line between edge of breakup and a full crunch. Combine that with the LDR pots that people in this thread are trying to implement, and you could have the most awesome super versatile programmable preamp that could go from tweed or blackface cleans to a Marshall roar to singing Boogie or anywhere between. Well, maybe not that versatile, as I can't think of a solution to the different cap values in various preamp flavors, but you'd still have a programmable amp that's far more versatile than the current crop of amps.

Now, I know there's people out there who are much smarter than me, and know more about electronics than I, and more than a few haunt these boards. Beyond what's beeen discussed in this thread thus far, what would be the considerations that would make using LDR variable resistors as digitally controllable plate and cathode resistors impractical?

[Steve Conner]

It would be quite practical, but load lines and bias points just don't make much of a difference to the tone, compared to the usual controls provided on a tube amp.

[Robert M. Martinelli]

It would be quite practical, but load lines and bias points just don't make much of a difference to the tone, compared to the usual controls provided on a tube amp.

Hmmm.....Steve, with all due respect, allow me to disagree on this one....
load lines and bias points DO have a great degree of influence on tone, because they affect a stage's gain and headroom and the way that stage might "clip", ( think about grid current limiting, cutoff, the possibility to introduce symmetric/asymmetric clipping which in turn can enhance odd/even harmonics etc. etc. ), so they allow the signal's waveform to be manipulated.

The actual point, IMHO, is that, of all the "possible" bias/load line settings, only a few are viable/effective/good sounding, and I think this is the main reason why no one offers the possibility to render these parameters variable by users....it's simply not worth the cost...

JM2CW

Cheers

Bob

[Steve Conner]

Yeah, I agree. I suppose they do influence the tone, but not in a very musical way, that's all.

I've built amps with adjustable bias points for preamp tubes, and there seemed to be one spot on the bias control that sounded right, and outside of that it sounded like an old transistor radio with a flat battery.

The cold biased stage in a SLO etc. would be one exception, I guess. But I'd simply have a switching arrangement to bypass that stage (and maybe a few others) for cleans.

[Koreth]

I've built amps with adjustable bias points for preamp tubes, and there seemed to be one spot on the bias control that sounded right, and outside of that it sounded like an old transistor radio with a flat battery.

The cold biased stage in a SLO etc. would be one exception, I guess. But I'd simply have a switching arrangement to bypass that stage (and maybe a few others) for cleans.

Which is why I was thinking of affecting load lines as well, there may very well only be a few bias points that sound ideal with each load line. Change the load line however, be it through a different plate resistor value or a different voltage to the plate from the supply and you'll probably have to change the bias point too to get it back to that point where that stage (and all others it effects) sounds ideal.

This, combined with a switching arrangement as you suggest to bypass gain stages would give you a way to stuff a lot of voicings into a single amp.